STATISTICS FOR SPATIO-TEMPORAL DATA

This is a master-level topic course on spatio-temporal modeling as part of the interdisciplinary activities of the Master's Degree Program in Computer Science ( Curriculum in Data Management and Analytics). Together with the other courses on statistical inference and learning and on statistical computation and simulation, it offers a spectrum of modern statistical techniques that nowadays represent the most required expertises. The major emphasis is on statistical models for discrete time data. Topics covered in the course will also include geostatistical models and spatial prediction. The focus is on applications with real data and their analysis with statistical programs such as R.

Produce informative visual displays of temporal and spatial data
Perform exploratory data analyses
Select appropriate statistical models for various types of temporal or spatial data
Estimate model parameters with modern statistical software ( R)
Interpret software output
Effectively communicate analysis in a written document and oral presentation

Probability and Mathematical Statistics at the level of Rice, J.A. (2007) Mathematical Statistics and Data Analysis, Duxbury Press

Time series analysis
- Filtering
- Decomposition
- Prediction
- ARMA Models
Spatial Markovian models and some generalizations
- Prediction

Shumway, Robert H., Stoffer, David S. (2017) Time Series Analysis and Its Applications: With R Examples. Springer
Banerjee, S., Carlin, B. P., & Gelfand, A. E. (2014). Hierarchical modeling and analysis for spatial data. 2nd Ed. CRC Press.
Additional material suggested by the teacher during the course.

Final examination will consist of two steps:
1) Preparation of an individual written report including the analysis of a dataset.
The class project will entail choosing a problem (mutually agreed upon), writing code to solve it, and writing a report which provides the background, motivation, methodology and
results. This project should ideally be related to the thesis so as to be multipurpose.
2) oral presentation (exam)
Final grades will be determined by the oral exam (50%) and the written project (50%)

This course is based on lectures, which will cover the major topics, emphasizing and discussing the important points.
Theoretical lectures will be complemented by exercise classes and lab sessions. The statistical software used in the course is mainly R (www.r-project.org).
Student's participation is essential, and reading the assigned material to reinforce the lectures will help the student to learn more efficiently.
R scripts from various sources may be used to reinforce the material.

English

oral